Finding Important Protein Motions in Solution Through Linear Dimensionality Reduction
Authors: Krzysztof Borowski and Forbes Burkowski.
Abstract:
Automating protein structure analysis is useful to understand their function, especially as structure data grows. Developing methods of elucidating protein motions from the multitude of data currently available, and doing it in a way that allows non-experts to easily perceive protein movement can lead to heightened levels of understanding and hypothesis generation. We apply principal component analysis to examine major modes of motion for protein conformations. We show that a few principal components of the conformation matrix can capture the majority of the motions, and present a visual depiction of one mode of calmodulin.